EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE: AN EVIDENCE BASED APPROACH

EVIDENCE BASED EMERGENCY MEDICINE: EVALUATION AND DIAGNOSTIC TESTING

073%3627/99 $8.00

+ .OO

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE: AN EVIDENCE BASED APPROACH Adam G. Field, MD, and Ernest Wang MD

Over 10 million patients per year visit their primary care physician or emergency department (ED) with a chief complaint of headache.39 Most of these headaches have a benign cause and require only a complete history and physical examination to make the correct diagnosis. There are, however, a small percentage of headaches that require urgent diagnostic studies and treatment. Headache may be the initial symptom of life-threatening causes such as subarachnoid hemorrhage (SAH), intracranial hemorrhage, mass lesions, encephalitis, meningitis, vasculitis, venous sinus thrombosis, and toxic metabolic disorders. There are many pressures placed on the emergency physician in the setting of managed care to diagnose and treat patients in the most economical manner possible. These pressures must be weighed, however, against the failure to diagnose and treat life-threatening illnesses, patient anxiety about a diagnosis based on a history and physical examination alone, and the medicolegal concerns of the treating physician. This article reviews the literature on the diagnostic tools available to make a cost-effective yet appropriate diagnosis in the patient with an atraumatic headache in the emergency setting. The tools that will be addressed include the use of a good history and physical examination alone, third-generation CT, lumbar puncture, and magnetic resonance imaging (MRI).The epidemiology and the characteristicsof the commoner

From the Division of Emergency Medicine, Northwestern University Medical School, Chicago, Illinois

EMERGENCY MEDICINE CLINICS OF NORTH AMERICA VOLUME 17 * NUMBER 1 * FEBRUARY 1999

127

128

FIELD & WANG

primary and secondary causes of headache are also reviewed, allowing the clinician to develop a better pretest probability of disease and make a more educated decision as to when additional diagnostic testing is needed. EPIDEMIOLOGY OF HEADACHES

In 1 year, over 70% of the population in the United States have a headache, and over 1%of physician’s visits and ED visits are for headache.48The cost to the United States from migraines alone, with regard to lost time from work and low productivity each year, is between $5 to $17 million.35A dramatic difference in health care utilization between migraineurs and a comparison group was demonstrated in a retrospective review of a managed care organization’s medical and pharmacy claims database. Total medical and pharmacy costs were $3.4 million for the migraine group and $2.1 million for the comparison group. The migraine group used $83,537 for diagnostic procedures compared with $13,140 for the comparison group.5 Various headache prevalence studies have been done in children, clinic populations, the general population, elderly patients, and HIVpositive patients. There are several problems with these epidemiologic studies. It is difficult to assess the validity of different headache classification systems, as there is no ”gold standard” nor are there biologic markers with which to compare specific diagnostic criteria.29There may also be interobserver variability in the collection of data and the clinical assignment of a diagnosis.’$ The present classification system of headaches may not apply to all populations. Most studies are retrospective chart reviews and are based on clinic or hospitalized patient populations that exclude a large percentage of the population who never seek care for their headaches. Rasmussen et a1 did the first prevalence study of headaches in the general population using the International Headache Society (IHS) classification system. One thousand 25- to 65-year-old men and women, who lived in the western part of Copenhagen County, were randomly drawn from the Danish National Central Person Registry. The lifetime prevalence of headache, including migraine, tension-type headache and all other types was 96% (709/740). The prevalence was higher in female (99%) than among male subjects (93%).The prevalence of tension-type headaches was 78% (578/740). Men had statistically significant lower prevalence than did women for tension-type headaches (69% versus 88°/0).39Similar data are reported in a recent epidemiologic study by Schwartz et a1 that showed women have a 16% higher incidence of episodic tension headaches than men Migraine headaches also have a gender preference, with a predilection toward women. In the Copenhagen study, the overall prevalence of migraine headache was 16% (119/740) with 8% being men and 25% being women.39The American Migraine Study also showed an increased

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

129

prevalence of female subjects, but it was more apparent during the reproductive years. The data from this as well as other studies demonstrate that before the age of 12 years, migraine is more common in male subjects, but after puberty it becomes more common in female subjects. At 20 years of age the ratio of female to male was about 2 : 1 and at 42 years it was 3.3:1.39 Age is another important epidemiologic factor to use in formulating a pretest probability for secondary causes of headache. Although most epidemiologic studies on headache have focused on younger people, headache is not an uncommon complaint of elderly. In one retrospective study of people over 65 years of age, 17% reported frequent headaches.6 The prevalence of headaches in the elderly in other studies range from 5% to 50%.20 Older patients have an increased incidence of coexisting medical conditions that may contribute to the increased morbidity of headaches seen in this population. Many studies have shown that the risk of headaches from serious conditions is increased dramatically in the elderly population. Pascual and Berciano retrospectively studied 3578 patients who presented to a neurology service with headache as their initial and main symptom. They concentrated on a 193 (5.4%) of the patients in whom the headaches occurred for the first time after the age of 65 years. Fifteen percent of the elderly patients versus one to two percent of patients under the age of 65 years, presented with headache secondary to serious conditions such as stroke, temporal arteritis or intracranial neoplasm. Although the overall incidence of patients with new headaches decreased with age, the risk of headache caused by serious secondary conditions increased after the age of 65 years. Tension and migraine headaches began more often in patients under the age of 65 years, whereas postherpetic trigeminal neuralgia and headache secondary to vascular and cervical disorders were significantly commoner in patients after 65 years.20 Other studies support age as a major historical factor in predicting the presence of intracranial pathology. Ramirez-Lopez et a1 reviewed the ED records of 139 hospitalized and 329 randomly selected patients from 1720 nonhospitalized adult patients to define clinical predictors of intracranial pathology independent of an abnormal neurologic examination. Besides an isolated abnormal neurologic examination, age greater than 55 years was one of the best positive predictors of intracranial pathology. Eleven of one hundred and one patients greater than 55 years had intracranial pathology versus seven of three hundred forty-nine under the age of 55 years.37This study did have some selection bias in that approximately 29.7% (139/468) of the patients in the study were hospitalized. In actuality, the hospitalized population only comprised 7.5'10 (139/1859) of the total patients studied. The hospitalized patients consist of a much higher percentage of elderly patients as compared with the entire population studied. The hospitalized patients in general also have a higher percentage of intracranial pathology. These factors

130

FIELD & WANG

could lead to slightly more dramatic conclusions as to the percentage of elderly patients with intracranial pathology in this study. HIV-positive patients are another population in which the prevalence of more serious secondary causes of headaches may be greater than in the general population. Headaches in this population are difficult to study because many different determinants may coexist in the same patient.

Headache in HIV-infected Patients2I Risk Factors Progressive immune deficiency leading to opportunistic infections and tumors High prevalence of depression and anxiety Alcohol, intravenous drug, or other substance abuse Multiple medications used in treating advancing disease15 Headaches at Different Stages of HIV Infection Headaches that can occur during all stages of HIV Acute aseptic meningitis Chronic headache and persistent pleocytosis Syphilitic meningitis Non-HIV-related headaches (eg., migraine, tension, sinusitis) Headache associated with more advance disease (CD, count less than 500 and/or 200) Opportunistic infections and tumors Meningitis (e.g., cryptococcal, tuberculous, syphilitic, and lymphomatous) Focal brain lesions (e.g., toxoplasmosis, CNS lymphoma, PML, abscess, cryptococomma) Diffuse brain lesions (e.g., CMV, HSV, and toxoplasmosis Medication use Zidovudine-induced headaches The setting of the study may also lead to very different results. For example, in a series of outpatient volunteers, 4% of the headaches were due to an opportunistic infection or tumor,5l compared with 82% when the population was defined as those patients presenting to an AIDS clinic or the ED.27 The prevalence of headaches in the HIV-infected population is not known on a large scale. There have been many studies, however, conducted on subpopulations of HIV-infected patients. In one prospective study assessing pain symptoms in an outpatient setting, 11%of asymptomatic seropositive persons at baseline complained of headache compared with 39% with ARC or AIDS.15 At the end of the first year, 16% of the previously asymptomatic subjects complained of new-onset headaches. In another retrospective chart review of hospital admissions, headache was present in 55% of patients with ARC or AIDS. Many of the studies show that the prevalence of headache is commoner than in the general population and increases in frequency as the disease progresses. The physician’s clinical suspicion for more serious secondary

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

131

causes should also increase as disease progresses and CD, count decreases. Headaches also occur commonly in children and adolescents. Headache may be due to the same primary causes in adults such as tension or migraine or may be due to secondary causes such as CNS tumors, systemic disorders, or infectious disease. Many studies have tried to estimate the frequency of headaches in the pediatric population. Bille studied about 9000 children in Sweden, 40% by the age of 7 years and 75% by the age of 15 years had already had a significant h e a d a ~ h eIn .~ another study, 82% of 13 year olds reported headaches in a 12-month period and 17% complained of more frequent headaches.42 As in adults, most headaches in children are benign in nature. The headaches that are common in adults like tension and migraine headaches are also common in children. In approximately one half of the patients with migraines, the attacks start before 20 years of age. A small percentage of headaches in children may be due to more serious secondary disease. Although less common in children than adults, cerebral aneurysms and intracranial bleeds can occur in this population. Infectious origins such as meningitis also may present initially as headache in children. Primary brain neoplasms are far more prevalent in children than they are in adults.41They account for almost 20% of all cancer in children, but only 1%in adults. CNS tumors are the second leading cause of cancer-related deaths in children under 15 years of age.50The diagnosis can sometimes be elusive and initially misdiagnosed as a viral syndrome or gastroenteritis in a young child. EPIDEMIOLOGY OF SUBARACHNOID HEMORRHAGE AND INTRACEREBRAL ANEURYSMS

The epidemiology of ruptured intracerebral aneurysms deserves special consideration because of the extremely high mortality associated with misdiagnosis. SAH from a ruptured aneurysm is probably one of the most devastating diagnoses to miss, especially in a young and otherwise healthy patients. The annual incidence of nontraumatic subarachnoid hemorrhages is approximately 25 in 100,000 or about 25,000 cases per year in the United States. Ten percent to fifteen percent of patients die before reaching the hospital, and fifty percent die within the first 6 months of diagnosis. The mean age of presentation is 50 years of age, and the disease occurs more frequently in women than men.13An estimated 18,000 of these patients are left severely disabled or dead.60 Other epidemiologic characteristics should increase a physician’s clinical suspicion for subarachnoid hemorrhage and unruptured intracranial aneurysms. Diseases such as polycystic kidney, coarctation of the aorta, abdominal aortic aneurysm, Ehlers-Danlos or Marfan’s syndromes, fibromuscular dysplasia, hereditary telangiectasia, Graves’ disease, and sickle cell disease are associated with an increased risk of SAH. Certain environmental factors such as cigarette smoking, athero-

132

FIELD & WANG

sclerotic heart disease, hypertension, and cocaine use should increase the physician's clinical suspicion as well. An increased incidence of familial aneurysms has also been demonstrated in many studies independent of other genetically related diseases.40A study by Schevink'et a1 determined the relative risk of patients with the familial form of SAH to be 4.14 compared with the general p o p ~ l a t i o n . ~ ~ Patients with unruptured aneurysms may also present with heada c h e ~ .A ~ ?retrospective analysis of 132 patients with unruptured aneurysms demonstrated that 54 patients had prior symptoms of the unruptured aneurysm.38Headache may be a warning sign of an impending aneurysmal SAH. The five largest series of patients with warning signs and intracranial aneurysms, which include 629 patients, show that about 50% of the patients with a bleeding aneurysm will have had a premonitory warning leak. Headache was a warning symptom in nearly all of the patients in the studies reviewed.34A patient may present with headache as well as other associated symptoms (discussed later) from days to months before a major intracranial bleed. CLASSIFICATION SYSTEM OF HEADACHES

In 1988, the IHS developed a new classification system for the diagnosis of 129 different types of headaches. Headache types can be described as primary or secondary. The primary causes include migraine, tension, and cluster headaches and account for about 90% of the headaches seen in the emergency setting. Primary headaches are usually benign but are often recurrent. Secondary headaches are often a symptom of an underlying disease process or other external factors. A discussion of all of the causes of secondary headaches is beyond the scope of this article. Some of the commoner secondary headaches seen in the emergent setting, as well as the secondary headaches with greater morbidity and mortality are addressed here. The IHS developed the following major categories for headaches:33

Major Categories of Headache Disorders Migraine Tension type Cluster headache and chronic paroxysmal hemicrania Miscellaneous headaches unassociated with structural lesion Idiopathic stabbing, external compression, cold stimulus, benign cough, benign exertional, associated with sexual activity Headache associated with head trauma Headache associated with vascular disorders Acute ischemic cerebrovascular disorder, intracranial hematoma, SAH, unruptured vascular malformation, arteritis, carotid or vertebral artery pain, venous thrombosis, arterial hypertension, associated with other vascular disorders Headache associated with nonvascular intracranial disorder

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

133

High and low cerebrospinal fluid pressure, intracranial infection, intracranial sarcoidosis, and other noninfectious inflammatory disease, related to intrathecal injections, intracranial neoplasm, associated with other intracranial disorder Headache associated with substances or their withdrawal Acute and chronic substance use or exposure, withdrawal after acute and chronic use, associated with substances or their withdrawal with uncertain mechanism Headache associated with noncephalic infection Viral infection, bacterial infection, other infection Headache associated with metabolic disorders Hypoxia, hypercapnia, mixed hypoxia and hypercapnia, hypoglycemia, dialysis, other metabolic abnormalities Headache or facial pain associated with disorder of cranium, neck, eyes, ears, nose, sinuses, teeth, mouth, or other facial or cranial structures Cranial neuralgias, nerve trunk pain, and deafferentation pain Persistent pain of cranial nerve origin, trigeminal neuralgia, glossopharyngeal neuralgia, nervus intermedius neuralgia, superior laryngeal neuralgia, occipital neuralgia, central causes of head and facial pain other than tic douloureux Migraine Headaches Migraine headaches are probably one of the commonest types of headache seen in the ED second only to tension headaches. An estimated 23.6 million people in the United States fit the IHS classification of migraine headache.55 The new classification system by the IHS had replaced the old terminology of classic or common migraine with migraine with or without aura. To be classified as a migraine without aura, the headache must have occurred at least five times with the following characteristic^:^^

I. Migraine without aura (Previously used terms: common migraine, hemicrania) Diagnostic criteria A. At least five attacks fulfilling B through D B. Headache attacks lasting 4 to 72 hours (untreated or unsuccessfully treated) C. Headache has at least two of the following characteristics 1. Unilateral location 2. Pulsating quality 3. Moderate or severe intensity (inhibits or prohibits daily activities) 4. Aggravation by walking stairs or similar routine physical activity D. During headache at least one of the following:

134

FIELD & WANG

1. Nausea or vomiting 2. Photophobia and phonophobia E. At least one of the following: 1. History, physical and neurologic examinations do not suggest one of the disorders listed in groups 5 to 11 (other causes of headache). 2. History and or physical and or neurologic examinations do suggest such disorder, but it is ruled out by appropriate investigations. 3. Such disorder is present, but migraine attacks do not occur for the first time in close temporal relation to the disorder. 11. Migraine with aura (Previously used terms: classic, ophthalmic, hemiparesthetic, hemiplegic, or aphasic) Diagnostic Criteria A. At least two attacks fulfilling IIB B. At least three of the following four characteristics: 1. One or more fully reversible aura symptoms indicating focal cerebral cortical or brainstem dysfunction 2. At least one aura symptom develops gradually over more than four minor two or more symptoms occur in succession. 3. No aura symptom last more than 60 minutes; if more than one aura symptom is present, accepted duration is proportionally increased. 4. Headache follows aura with a free interval of less than 60 minutes (it may also begin before or simultaneously with the aura). C. At least one of the following: 1. History, physical and neurologic examinations do not suggest one of the disorders listed in groups 5 through 11. 2. History or physical or neurologic examinations do suggest such disorder, but it is ruled out by appropriate investigation. 3. Such disorder is present, but migraine attacks do not occur for the first time in close temporal relation to the disorder. 111.Migraine with typical aura33 Diagnostic Criteria A. Fulfills criteria for migraine with aura including all four criteria under B B. One or more aura symptoms of the following types: 1. Homonymous visual disturbance 2. Unilateral paresthesias and/or numbness 3. Unilateral weakness 4. Aphasia or unclassifiable speech difficulty

Migraine with and without aura initially begins prior to 40 years of age. Eighty percent of migraineurs do not have aura, but some of these patients may have a premonition of onset of headache. These patients may have a vague prodrome including lethargy, depression, hyperactivity, and food craving.I4When an aura occurs it typically starts 1 hour

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

135

before the onset of headache and is completely reversible. Aura symptoms associated with migraine include the following: 1) scotoma (blind spots); 2) teichopsia (bright wavy lines); 3) fortification (zigzag pattern); 4) Photopsia (flashing lights); 5) visual and auditory hallucinations; 6) paresthesias; 7) metamorphopsia (distorted size of objects). The symptoms of aura or the temporary neurologic deficits associated with complicated migraines (hemiplegic, ophthalmoplegic, and basilar) often are the reason why clinicians do further diagnostic evaluation. A better history may help avoid some of these tests. Precipitating factors also may help to differentiate migraine headache from other more serious secondary causes. In women, there is an increased incidence of migraines before, during, or immediately after their period. Certain foods containing tyramine (red wine and cheese), amines (chocolates), nitrates, MSG, and alcohol may precipitate migraines. Caffeine withdrawal may have a similar effect. Stress, changes in weather, or changes in sleep patterns may also precipitate migraines.I4 The IHS criteria for the diagnosis of migraine headaches are not perfect in all settings. In the pediatric population, Winner et a1 proposed revisions of the criteria including duration of 1 to 48 hours, location either bifrontal/bitemporal or unilateral, and symptoms that include photophobia or phonophobia. Eighty-eight children were studied prospectively at eight specialty clinics comparing the IHS criteria and the revised criteria. They reported a 93% sensitivity with the revised criteria versus 66 percent sensitivity with the IHS criteria in the clinical diagnosis of migraine in children.61Solomon and Lipton also proposed simplified criteria to increase sensitivity of diagnosing migraine in clinical practice. They suggested that the original criteria were too complex and overly restrictive. They proposed that criteria for migraine without aura should include any two of the following: headache-unilateral site; headache-pulsating quality; associated nausea; and associated photophobia or phonophobia, along with other criteria, specifically similar headaches have occurred in the past, and no evidence of organic disease.54A major problem in all of these studies is that there is no real gold standard to validate any of these diagnostic criteria. Solomon and Lipton suggest that the strict criteria of the IHS may exclude some of the migraineurs who may benefit from current migraine therapy. Some clinicians have tried to use therapy as a diagnostic option for migraines. Several case reports suggest that this option is not an appropriate diagnostic tool. Seymour et a1 report three cases in which headaches completely resolved with nonnarcotic agents and were therefore presumed to be benign in nature. All three patients were subsequently found to have intracranial Another case report by Lipton et alZ6demonstrated that sumatriptan relieves migraine-like headaches associated with carbon monoxide exposure. Relief of the headache with medication, therefore, does not rule out the possibility of life-threatening causes.

136

FIELD & WANG

Tension Headaches

Tension headaches are one of the commonest headaches seen in the ED and the primary care setting. Tension headaches were previously termed muscle contraction headache, psychomyogenic headache, stress headache, ordinay headache, and psychogenic headache. The IHS divides tension headaches into episodic and chronic, and also based on the presence or absence of pericranial muscle t e n d e r n e ~ s . ~ ~ I. Episodic tension-type headache Diagnostic Criteria: A. At least 10 previous headache episodes fulfilling criteria B through D listed below; number of days with such headache <180/yr (< 15/month) B. Headache lasting from 30 minutes to 7 days C. At least two of the following pain characteristics: 1. Pressing/tightening (nonpulsating) quality 2. Mild or moderate intensity (may inhibit but does not prohibit activities) 3. Bilateral location 4. No aggravating by walking stairs or similar routine physical activity D. Both of the following: 1. No nausea or vomiting (anorexia may occur) 2. Photophobia and phonophobia are absent, or one but not the other is present 11. Chronic tension-type headache (previously termed chronic daily headache) Diagnostic Criteria: A. Average headache frequency 15d/mo (180 days/yr) for 6 months fulfilling criteria B through D B. At least two of the following pain characteristics: 1. Pressing/ tightening quality 2. Mild or moderate severity (may inhibit but does not prohibit activities) 3. Bilateral location 4. No aggravation by walking stairs or similar routine physical activity C. Both of the following: 1. No vomiting 2. No more than one of the following: nausea, photophobia, or phonophobia D. At least one of the following: 1. History, physical and neurologic examinations do not suggest one of the disorders listed in groups 5 through 11 2. History or physical or neurologic examinations do suggest such disorder, but it is ruled out by appropriate investigations 3. Such disorder is present, but tension-type headache does not occur for the first time in close temporal relation to the disorder

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

137

In general, a tension-type headache is usually described as pressing or tightness that is diffuse and sometimes associated with pericranial muscle tenderness. It is usually moderate in intensity, not severely disabling, and not associated with the typical symptoms of migraine such as nausea, vomiting, and photophobia. The precipitating factor is often related to stress. If the headaches are present greater than 15 days a month for longer than 6 months they are considered chronic rather than episodic. Cluster Headache

Cluster headaches are severe, excruciating, sharp unilateral headaches that usually last from 30 to 90 minutes and occur one to two times daily for several weeks at a time. There may be intervals of months to years in which the patients are symptom free. The headaches typically occur at the same time each day, often awakening the patient from sleep. Unlike migraines, there is no prodrome. Associated symptoms include conjunctival injection, ptosis, miosis, tearing, nasal congestion, and forehead sweating of the affected side. The headaches typically occur between 20 and 40 years of age and are more common in males. The patients may also report that the headache attacks are precipitated by

OTHERSELECTEDHEADACHEDISORDERS

Some of the commoner secondary headache causes seen in the ED include sinusitis, dehydration, systemic infections, carbon monoxide (CO) exposure, cranial neuralgias (postherpetic neuralgia), and postlumbar puncture headache. The secondary headaches with an extremely high morbidity and mortality include SAH, meningitis, glaucoma, stroke, temporal arteritis and brain tumors. Sinusitis

Sinusitis is often overdiagnosed in the ED. The patient typically presents with sinus congestion, tenderness to palpation over the sinuses, and worsening of head pressure with leaning forward. Although headaches from sinusitis are usually located over the sinuses, they may also be referred to other areas of the head or neck. The symptoms of sinusitis may be divided into major and minor factors. Major factors are facial pressure and pain, facial congestion and fullness, nasal obstruction and blockage, nasal drainage, and postnasal drip. Minor factors are headache, fatigue, halitosis, fever, dental pain, cough, and ear pressure. Sinusitis may be acute, subacute, or chronic. The cause may be secondary to infection, allergens, anatomic obstruction, or b a r ~ t r a u m a . ~ ~

138

FIELD & WANG

lntracranial Hypotension

Systemic infection, dehydration, hyperpnea, profound hyperglycemia and post-lumbar puncture typically cause headache by creating a relative intracranial hypotension. The typical features include pain that is aggravated by upright position and relieved with recumbency. Head shaking also aggravates pain. Associated symptoms include orthostatic lightheadedness, tinnitus, photophobia, anorexia, and nausea and vomiting. Post-lumbar puncture headache secondary to CSF leak has been reported to be a complication of the procedure in 5% to 30% of procedure~.~~ Postherpetic Neuralgia

One of the commoner cranial neuralgia seen in the ED setting is herpetic or postherpetic neuritis of the trigeminal nerve. The pain of herpes zoster is sustained in contrast to the paroxysmal pain of trigeminal neuralgia. The pain usually regresses 2 to 3 weeks after the initial herpetic outbreak, but it may persist for months to years. The pain is in the distribution of the fifth cranial nerve and also involves the forehead. Initially there will be hyperesthesias in a unilateral V1 distribution followed by the appearance of erythematous and herpetiform lesions. Chronic scarring, pain, and paresthesias may persist weeks to months after the herpetic eruption has resolved. This disease entity is commoner in the elderly. Other secondary causes of cranial neuralgias should be ruled out with further diagnostic testing if herpetic or postherpetic neuralgia is not obvious by history and physical examination.* Carbon Monoxide Exposure

Mild CO poisoning is associated with nonspecific headaches and flulike symptoms. Greene et a1 measured CO levels in 100 patients with non-specific headaches who registered in the emergency department between November 1980 and March 1981. Patients with a history suggestive of carbon monoxide exposure were excluded. The authors found A clinithat there was a 7% prevalence of unsuspected CO expo~ure.'~ cian should have a high index of suspicion for CO poisoning a patient with nonspecific headache and vague viral symptoms, especially if many other members of the household have the same symptoms. Eye Pain and Headache

Headaches that present as a painful eye or retro-orbital pain may be secondary to ocular disease or other intracranial processes. Daroff stated that "a white eye is not the cause of a painful eye."56If the eye

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

139

looks normal on physical examination for the most part, the pain is probably referred from another intracranial process. The sensory innervation of the eye is from the first branch of the trigeminal nerve; other branches of the trigeminal nerve supply the dura, intracranial vessels, and venous sinuses. This allows other intracranial processes such as aneurysms, tumors, and venous sinus thrombosis to present with pain referred to the eye. Eye pain secondary to ocular causes can be due to eyestrain, corneal abrasions, glaucoma, uveitis, iritis, ischemic ocular pain, orbital cellulitis, orbital pseudotumor, and orbital or retroorbital mass lesions. Eyestrain, corneal abrasions, and uveitis/iritis are easy to diagnose by a good history and physical examination. The emergency physician should be alert to the presentation of glaucoma. Acute-angle glaucoma is a rapid rise in intraocular pressure and is often associated with acute pain and headache. The pain is often confused with migraine or intracranial aneurysm. The pain is around the orbit and often is associated with nausea and vomiting. Occasionally a patient may present with abdominal pain and vomiting. The eye is red, the pupil is often midrange, and visual acuity is decreased. If the diagnosis is missed, acute glaucoma may result in b l i n d n e ~ s . ~ ~ Temporal Arteritis Temporal arteritis or giant cell arteritis is the commonest vasculitis affecting older people. According to the American College of Rheumatology criteria, the diagnosis of temporal arteritis requires the presence of three of the following five criteria: Age greater than 50 years; new onset of localized headache; temporal artery tenderness of decreased pulse; erythrocyte sedimentation rate at least 55 mm/hr; and positive histology on biopsy. There is also an increased incidence of temporal arteritis associated with polymyalgia rheumatica. Other associated symptoms include jaw claudication, amaurosis, anorexia, weight loss, muscle aches, transient ischemic attacks, neuropathies, and stroke. Headache is the commonest symptom of temporal arteritis. The pain is severe, but may be sharp or dull. Tenderness of the temporal artery is present in about one half the patients. In some studies, temporal headaches are described 25% to 50% of the time. In 28% the headache did not involve the temporal area. Failure to diagnose and treat temporal arteritis can result in blindness.*O Brain Tumors Headache is a common manifestation of increased intracranial pressure. Causes of increased intracranial pressure include mass lesions (e.g., brain tumors, abscesses, chronic subdural hematomas), pseudotumor, and idiopathic intracranial hypertension. Headaches from brain tumors are typically aching and dull, become progressively worse over time,

140

FIELD & WANG

and sometimes are worse in the morning. Valsalva’s maneuver and coughing may increase the headache pain by increasing intracranial pressure.49Focal neurologic deficits or seizures are often present. Not all brain tumors present with classic symptoms, however. Some patients present with only very mild headaches. There are some case reports in the literature of headaches from brain tumors presenting with the typical symptoms of a migraine. Brain tumors often present with other signs and symptoms that may give the physician an indication to do further diagnostic testing. In one prospective study of the clinical manifestations of patients suffering from intracranial tumors, only 8.2% exhibited headache as their first and isolated clinical symptom. Focal signs and symptoms occurred in 57% of the cases, and seizures in 9%. If the isolated headache had a longer than 10-week duration without other associated symptoms, it was much less likely to be due to intracranial Many retrospective chart reviews have been done to determine the presenting signs and symptoms of patients with brain tumors. Snyder et a153reviewed the presenting characteristics of 101 patients with a discharge diagnosis of primary brain tumor. All of these patients had initially been admitted through the ED. The commonest presenting symptoms (and number of patients with that symptom) were headache (56), altered mental status (51), ataxia (41), nausea or vomiting (37), seizure (24), visual change (23), speech deficit (21), and sensory abnormalities (18). The commonest presenting signs (and number of patients with that presenting sign) were motor weakness (37), ataxia (37), papilledema (28), cranial nerve palsy (26), visual change (20), sensory abnormalities (19), Babinski reflex (18), and speech deficit (12). Unfortunately, the lack of any of these signs and symptoms does not completely eliminate the possibility of a mass lesion. The presence of a combination of any of these symptoms should increase the clinician’s pretest probability of an intracranial process. Subarachnoid Hemorrhage The first diagnosis to consider when a patient presents to the ED with an acute-onset of the “first or worst” headache is SAH. In the classic presentation of a SAH secondary to a ruptured aneurysm, the patient complains of the acute onset of ”the worst headache ever,” associated with nausea and vomiting, photophobia, neck stiffness, and syncope. Fontanarosa retrospectively studied the presenting characteristics of 109 patients with SAH. Headache was present in 74% of patients, 77% had nausea and vomiting, and 53% had loss of consciousness. Nuchal rigidity was present in 35% of the patients, and 64% had some other neurologic findingsx6Headaches may present with similar or less severe symptoms and be a warning symptom of an impending rupture of an intracranial aneurysm. These are often referred to as sentinel bleeds, which are almost always a retrospective diagnosis. They are often

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

141

misdiagnosed as sinusitis, flu, tension headache, or muscular strain. Ostergaard reviewed the five largest retrospective studies of patients (629 total) with warning signs of intracranial aneurysms, and found that approximately 50% of patients admitted to neurosurgical intensive care units had some symptoms of a warning leak. Headache occurred in almost all of the patients (86%-97%). Other symptoms included nausea and vomiting, meningismus, visual disturbances, motor and sensory disturbances, photophobia, and loss of consciousness. All of these associated symptoms individually occurred in less than 30% of the patients, but greater than 60% had more than one associated Not all patients that present with the sudden onset of ”the worst headache of his/her life” have an SAH. Wijdicks et a1 followed up patients that were given a provisional diagnosis of SAH and who had an initial negative CT and lumbar puncture. Patients who had an angiogram during the initial presentation were excluded. The criteria for inclusion were a headache starting within seconds, reaching maximal intensity within 1 minute, and normal CSF fluid within 48 hours of the onset of headache. The mean follow up was 3.3 years. None of the patients in this retrospective study had SAHs in follow up. This headache was given the term thunderclap headache.62Other vascular and nonvascular disorders should be considered with the acute onset of ”worst” headache, but only after the most life-threatening diagnoses should be excluded. HISTORY, PHYSICAL EXAMINATION, AND HEADACHE DIAGNOSIS

Most of the studies discussed in the epidemiology of headaches and the classification of headache sections are retrospective or level I11 evidence. Even in the retrospective studies, most of the results do not have sufficient sensitivity or specificity to completely rule out intracranial pathology. Several historical and physical findings may increase the yield of diagnostic studies. The following scenarios should warrant further diagnostic testing:** ”First or worst” headache Increased frequency and increased severity of headache New-onset headache after 50 years of age New-onset headache with a history of cancer or immunodeficiency Headache with mental status changes Headache with fever, neck stiffness, and meningeal signs Headache with focal neurologic deficits if not previously documented as a migraine with aura Other historical features such as family history, associated signs and symptoms, and associated risk factors may increase pretest probability of certain disease processes. The lack of the historical evidence or physical findings does not rule out serious intracranial pathology, however.

142

FIELD & WANG

EVALUATION OF DIAGNOSTIC TESTING EVIDENCE Search Strategy and Evidence Assessment

The authors performed a literature search on MEDLINE through December 1997. Keywords were searched including headache, computed tomography, subarachnoid hemorrhage, magnetic resonance imaging, brain neoplasms, and migraine. Also searched were textwords such as new-onset headache, nontraumatic headache, and chronic headache. From selected articles obtained from the MEDLINE search, the authors performed a bibliography review for additional papers. Prospective, retrospective, descriptive studies, and reports of expert committees were included. Case series and reports were excluded. The methodologic qualities of the articles were rated on the following scale: Level I: Evidence obtained from well-designed randomized controlled trials or evidence obtained from well-designed meta-analysis of randomized controlled trials Level 11: Evidence obtained from well-designed clinical studies, such as controlled trials without randomization, case-control studies, cohort studies, time series, or uncontrolled prospective studies Level 111: Evidence obtained from descriptive studies, case series or case reports, studies using nonrandomized, historical controls, or reports of expert opinion or expert committees HEADACHE TYPES New-onset Headache

Level I

There are no prospective, randomized controlled trials evaluating the use of neuroimaging in new-onset headache. Level I1

A recent prospective population-based study examined the diagnostic role of neuroimaging in patients with recent-onset heada~hes.'~ Over one-year, 100 consecutive patients who were referred for recent-onset headache to the neurology unit were included in the study. Eighty percent of the patients had normal neurologic examinations. Recent onset was defined as a headache that had persisted for at most 12 months. All patients underwent CT scanning with and without contrast. LP, ESR, MRI, MRA, and temporal artery biopsy were performed in selected clinically indicated patients. Tumors were diagnosed in 21% of the patients and composed 16% of the patients with normal neurologic examinations. The yield of CT overall in this group was 22.5%. A small prospective study by Liedo et a1 examined the relationship

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

143

between acute headache of recent onset and SAH.25Over the course of 1 year, all patients who were seen in the ED with acute headache of sudden onset and high severity and with no previous history of headache, obvious cause, or overt focal symptoms and signs on neurologic examination were included. All patients were examined with a thirdgeneration CT scan, with a mean headache duration of about 72 hours in SAH cases. Of the 27 patients entered into the study, 20 had normal scans and 4 were diagnosed with SAH. One frontal meningioma, a hyperdense lesion (hematoma associated with SAH), and diffuse enhancement (bacterial meningitis) were also found. Lumbar puncture was performed in 19 of the patients with normal scans, yielding five additional cases of SAH. Thus, CT scan demonstrated SAH in only four out of nine cases. In total, 9 cases of SAH were diagnosed out of 27 patients with new-onset headache, providing a 33% yield. Level 111

In a recent retrospective study of 1111 patients who underwent CT scans for headache, 120 (10.8%) abnormalities, including hemorrhage, infarct, or tumor, were diagnosed.23Only patients with acute headache or migraine were included, but migraine accounted for only 3.7% of the study group at one center. All studies were performed at two teaching hospitals over a 3-year period. There was significant variation in the percentage of intracranial abnormalities based on the setting in which the study was obtained. The inpatient rate (21.2%) was twice that for ED patients (11.7%) and three times that for outpatients (6.9%) ( P <0.005); 155 examinations were performed for headache as the sole presenting symptom (13.9%). Of these, nine (5.8%) patients had acute intracranial abnormalities. Summary of Evidence

Although there have been no prospective double-blind randomized controlled studies evaluating the utility of neuroimaging or lumbar puncture in new-onset headache, the available data support with moderate clinical certainty that some type of evaluation is indicated. The data presented show that 10% to 33% of patients presenting with new-onset headache have serious intracranial pathology. Studies focusing on patient populations presenting to the ED are lacking; however, Kahn et a1 demonstrated that ED patients have a higher incidence of serious abnormalities. CT scans have generally been the standard of care for the initial ED evaluation of new-onset headache because CT is quicker to obtain than MRI and less invasive than lumbar puncture. Until further data are available to demonstrate superior sensitivity of MRI, we recommend routine CT evaluation of all patients who present with new-onset headache, especially if abnormal neurologic findings are present and for those patients in whom the suspicion for SAH is high and the CT scan is nondiagnostic, an lumbar puncture should be performed. There are

144

FIELD & WANG

only isolated case reports of missed SAH with a negative CT and lumbar puncture.1° Chronic or Migraine Headache

Level I and II There are no prospective randomized controlled trials or evidence from meta-analysis from randomized controlled trials to support or refute the use of neuroimaging in patients with chronic headaches or migraine headaches. Level 111

In 1994, the American Academy of Neurology provided a summary statement on the use of neuroimaging in patients with headache and a normal neurological examination based on a review of the 1iteratu1-e.~~ They concluded that routine imaging ”in adult patients with recurrent headaches that have been defined as migraine-including those with visual aura-with no recent change in pattern, no history of seizures, and no other focal neurologic signs or symptoms . . . is not warranted.” This position is based on a 1994 literature review by Frishberg of 17 papers between 1974 and 1991 that were limited to studies with greater than 17 patients per study.17All patients had normal neurologic examinations. Of 897 CTs or MRIs done in migraine patients, only 3 tumors and 1 AVM were noted, resulting in a yield of 0.4% pathology; however, ”in patients with atypical headache patterns, a history of seizures, or focal neurological signs or symptoms, CT or MRI may be indicated.” In a recent retrospective study, 363 consecutive patients with chronic headache were evaluated by CT scan with and without IV contrast. All patients were referred from neurologists, neurosurgeons, primary care physicians, or specialists. None had CT scans in the ED setting and all had normal neurological examinations. The patients were divided into three groups based on their CT result. Group I (88.4%)had no abnormalities on CT scan. Group I1 (8.6%) had abnormalities considered not significant in the setting of chronic headache and for which no treatment was performed. Group I11 consisted of 11 patients (3%) who all demonstrated space-occupying lesions. Two intraventricular cysts and four meningiomas were found, but no intervention was undertaken. The rest required radiation and chemotherapy or surgery. IN total, 5 (1.4%)of the 363 patients with chronic headache had underlying pathology requiring intervention. Their conclusion was that even though the percentage of intracranial abnormality is small, it is not negligible and thus warrants use of CT scan in chronic heada~he.~ Mitchell et a1 conducted a prospective study on 350 consecutive patients presenting with a chief complaint of headache to determine the likelihood of a positive head CT scan.3oThe patients consisted of two

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

145

groups: active young military men and dependents of active duty personnel and retirees. Of these, 36% of the referrals for CT came from neurology department, 27% came from the ED, and 37% from specialty and primary care clinics. No reference to the type of headache or duration of headache was mentioned, however. Of the 350, 142(42%)were with contrast, 128 (37%) were without contrast, and 75(21%)were with and without contrast. Of 350 patients, 7(2%) had clinically significant findings on CT scan, including metastatic lung cancer, left frontal sinusitis with epidural abscess, meningioma, air in dural venous sinus, subacute rebleed into a subdural hematoma, noncommunicating hydrocephalus, and absence of cavum septum pellucidum. All had either abnormal physical findings or unusual clinical histories. Only 1 of 27 patients presenting with the worst headache of his/her life had a positive scan. Twenty-seven (8%) patients of the remaining 343 negative scans presented with abnormal neurologic or physical findings. Twenty-five (7%) had positive CT findings that were described as clinically insignificant (i.e., cerebellar atrophy, sinus disease, arachnoid cyst, lacunar infarction, venous angioma, basal ganglia calcification, and assymetric quadrigeminal cistern). Based on these findings, the authors conclude that the likelihood of discovering significant intracranial disease in a patient with normal examinations and no unusual symptoms is low. MRI has been shown to be more sensitive in detecting white matter changes in migraine patients. De Benedittis et a1 conducted a case series involving 63 consecutive patients with either migraine or tension-type headaches who underwent MRI scan to determine the incidence of cerebral abnormalities." These patients were compared with a group of 54 headache-free subjects of similar age. All patients had normal neurologic examinations. Focal white matter abnormalities on T,-weighted imaging were demonstrated in a significantly higher proportion of the headache sufferers than in the age-matched controls (33.3% vs. 7.4%) when the scans were reviewed by two independent neuroradiologists who were unaware of the source (patient vs. control). A retrospective review of MRI in 185 consecutive patients diagnosed with migraines, however, revealed a more modest percentage of patients with white matter abnormalities (30/185 or 16Y0).~ Statistically significant differences between the frequency of white matter abnormalities and patient age (>50 years) and presence of medical risk factors, including hypertension, diabetes, demyelinating disease, or autoimmune disorder was found (P <0.0001). Summary of Evidence The preponderance of evidence suggests that there is no need to obtain a CT scan or an MRI in patients with chronic headache or migraines in whom a typical pattern of intensity is described and who has an intact neurologic examination. These conclusions are based on level I11 evidence and expert opinion and therefore cannot conclusively be thought to recommend that neuroimaging be totally excluded in the

146

FIELD & WANG

work-up of a patient with a typical chronic headache. Furthermore, although some data support the superior sensitivity of MRI over CT, there are no data that demonstrate that the information gleaned from these examinations will change clinical management. Although the probability that an abnormality will be found is low, there will still be some level of diagnostic uncertainty that many physicians and patients will not be able to accept. In these cases, many physicians should err on the side of caution. Subarachnoid Hemorrhage

Level I

There are no prospective, randomized double-blind controlled studies evaluating the necessity of lumbar puncture if the head CT is negative. There are no studies to support or disprove that MRI is more sensitive than CT scan in acute SAH. Level I1

With the advent of new-generation CT scanners and increased experience with MRI, many studies have addressed the accuracy of CT or MRI in the diagnosis of SAH. Earlier studies have demonstrated that CT scans done within the first 2 days of the ictus can be 100% s e n ~ i t i v e . ~ ~ Detection decreases to 85% after 5 days, 50% after 1 week, 30% after 2 weeks, and almost undetectable at 3 weeks. These percentages were similar to those reported in the Cooperative Aneurysm Study, in which intracranial blood was detected in 95.3% on day 0, 90.5% on day 1, and 73.8% on day 3 after SAH.' Van der Wee et a1 conducted a consecutive case series on 175 patients who presented with sudden headache and normal neurologic examination^.^^ All patients had a noncontrast head CT scan done by a third-generation scanner within 12 hours of onset. Patients in whom the CT scan was normal subsequently had lumbar punctures at least 12 hours after the onset of the headache. Of the 175 patients, 117 had positive CT scans. Of the remaining 58 patients with normal scans, 2 were found to have SAH by LP. The detection rate of SAH by CT scan was 117 out of 119, or 98% sensitivity. They conclude that even if a patient presents with headache suspicious for SAH within 12 hours of onset, a lumbar puncture is still indicated if the initial head CT is nondiagnostic. As reported earlier, Liedo et a1 prospectively examined the relationship between acute headache of recent onset and SAH. Lumbar puncture was performed in 19 of the patients with normal scans, yielding five additional cases of SAH.25Thus, CT scan demonstrated SAH in only four of nine cases (44'/0), which is considerably less than previously reported in larger studies.

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

147

Level 111

A retrospective descriptive study was performed by Sames et a1 to determine the sensitivity of new-generation CT (NGCT) scans for the detection of acute nontraumatic SAH.43A review of the medical records from four neurosurgical centers was performed. Of these, 181 patients underwent NGCT scanning either within 24 hours of presentation (group 1)or after 24 hours (group 2) of symptom onset. The sensitivities of the NGCT scan for SAH were 93.1% and 83.8% respectively giving an overall sensitivity of 91.2%. The difference in sensitivity between the two groups was insignificant ( p CO.08). Sidman et a1 retrospectively reviewed 140 cases of diagnosed nontraumatic SAH evaluated by third-generation CT scans.47They concluded that the sensitivity was 100% (80/80) in patients imaged <12 hours after onset of headache, but only 81.7% (49/60) after 12 hours of symptom duration (P <0.0008). Of the 140 patients who had a negative CT, 11 had positive spinal fluid, giving an overall sensitivity was 92.1% (129/ 140). It is not clearly stated whether those patients with positive CT scans had the SAH confirmed by lumbar puncture, however. Several case studies have recently attempted to establish a superiority of MRI over CT scans in the diagnosis of SAH. Ogawa et a1 performed both head CT without contrast and MRI scan of the brain in 37 patients with known SAH diagnosed by lumbar puncture to determine which was more sensitive in detecting SAH.31Both studies were performed in all patients within 24 hours of the diagnosis by LP. Patients were divided into subacute (4-14 days after rupture) or chronic (>14 days). SAH was detected in 59% of subacute and only 10% of chronic SAH (46% overall). MR was able to determine subacute SAH in 25% to 89%, depending on the technique (TI-weighted,proton density, moderately T,-weighted, or T,-weighted images) and 25% to 100% of chronic SAH. The same group reported similar results in a similar study in which the neuroradiologists were blinded whether the scans performed were from patients with SAH or from patients who had normal MR scans for vertigo and headache.32 Matsumurua et a1 studied 27 consecutive preoperative and 33 postoperative patients with cerebral aneurysms, and compared MR with CT.28T,-weighted images detected blood in the basal cisterns and subarachnoid spaces in 25 of 27 patients, whereas CT scan confirmed SAH in 22 patients. They conclude that MR imaging was superior to CT scan for demonstrating blood in the ventricles or posterior fossa subarachnoid space. Summary of Evidence In the era of third- and fourth-generation CT scanners, the sensitivity of CT scan still does not approach 100% sufficiently to obviate the need for a lumbar puncture to rule out SAH. Indeed, seven of ten

148

FIELD & WANG

patients in Sames’s study with normal scans were found to have aneurysms by angi~graphy.~~ The authors recommend CT as the initial test for the diagnosis of SAH because of its proved sensitivity and the availability and expediency with which it can be obtained compared to MRI. The three recent studies comparing MRI with CT suffer from many methodologic deficiencies, including small numbers, lack of true randomization (because all patients were known to have SAH prior to MR), and lack of a sufficient number of reader-blinded evaluations of the MR scans. There is no level I evidence to support or disprove that lumbar puncture is necessary to rule out SAH if the head CT is negative. There is no level I evidence to support or disprove that MRI is more sensitive than CT scan in acute SAH. It is doubtful that any prospective doubleblind controlled studies will be able to be performed to evaluate the sensitivity of CT or MRI scanning without lumbar puncture because of the extremely high morbidity and mortality associated with missing an SAH. Based on the level I1 and 111evidence presented, lumbar puncture remains the test of choice in the ED when the initial CT or MRI is negative or nondiagnostic. FUTURE STUDIES

To better delineate which patients in the ED setting require emergent diagnostic testing, future studies should focus on large prospective reader-blinded trials and cohort studies designed to validate or refute the current body of evidence regarding indications for CT, MRI, and lumbar puncture. The incidence of intracranial abnormalities in patients presenting to the ED can be better defined by performing prospective studies whereby nonselected patients presenting to the ED with headache undergo CT scanning with new-generation scanners. Prospective cohort studies comparing the historical and physical findings in patients with CT abnormalities and in those with normal head CT scans can further clarify which characteristics of patients with chronic or acute headache are at risk for tumors or bleeds. This information can help physicians estimate the pretest probability of finding an intracranial abnormality that in the long run will result in more appropriate test ordering and cost-effectiveness. A relevant question that could also be addressed prospectively is the issue of whether neuroimaging is necessary if the patient’s headache improves with treatment in the ED, and if so, which tests are most sensitive? With respect to SAH, prospective studies need to be performed to determine which time interval from onset of headache to performance of a diagnostic CT scan will have a high enough sensitivity to obviate a lumbar puncture and to delineate also, which historical or physical examination findings stratify patients into a significantly higher-risk category for SAH. Further

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

149

studies also should address the role of angiography in diagnosing SAH and its sensitivity compared with lumbar puncture. The sensitivity of CT and MRI could be better evaluated by performing a study in which the patients have both studies performed but then are randomized to have either the CT or MRI read by a blinded neuroradiologist. Results would be compared to lumbar puncture for confirmation of the diagnosis, ensuring that an independent, blind comparison with a reference standard is performed. CONCLUSION

Nontraumatic headaches represent one of the commonest complaints in the ED. Most headaches are of benign origin and do not require emergent imaging or intervention. The authors' review of the diagnostic testing modalities does not offer absolute indications for neuroimaging because the conclusions are based mostly on level I11 and some level I1 evidence; however, we recommend further diagnostic testing when the headache is of new onset, has atypical characteristics, is of unusual severity, or is associated with focal neurologic deficits. New-generation CT scanning remains the initial ED test of choice for new-onset headache, headache suspicious for SAH, and headache suspected to be secondary to an intracranial mass or intracerebral bleed. The authors also recommend performing lumbar puncture when the CT scan is equivocal in ruling out SAH. The use of MRI in SAH will be institution dependent because the current evidence is inconclusive regarding its sensitivity as the primary imaging modality. Clinical diagnosis will always be paramount in the evaluation of headaches; however, for the small subset of patients who present with headache secondary to an intracranial mass, bleed, or SAH, making the diagnosis is crucial to prevent devastating morbidity and mortality. CT, MRI, and lumbar puncture all play prominent roles in the evaluation of these headaches, but their roles will continue to evolve as technology becomes more sophisticated and physicians become more expert with their use. References 1. Adams HI', Kassell NF, Tomer JC, et a1 CT and clinical correlations in recent aneurysmal subarachnoid hemorrhage: A preliminary report of the Cooperative Aneurysm Study. Neurology 33:981-988, 1983 2. Akpek S, Arac M, Atilla S, et al: Cost-effectiveness of computed tomography in the evaluation of patients with headache. Headache 35:228-230,1995 3. Atlas SW MR imaging is highly sensitive for acute subarachnoid hemorrhage . . . not! Radiology 186:319-322, 1993 4. Bille B: Migraine in school children. Acta Paediatr Scand Eil(supp1 136):l-151, 1962 5. Clouse JC, Osterhaus J T Healthcare resource use and costs associated with migraine in a managed healthcare setting. Ann Pharmacother 28:659-664, 1994 6. Cook N, Evans DA, Funkenstein H, et al: Correlates of headache in a populationbased cohort of elderly. Arch Neurol46:133&1344, 1989

150

FIELD & WANG

7. Cooney BS, Grossman RI, Farber RE, et al: Frequency of magnetic resonance imaging abnormalities in patients with migraine. Headache 36:616-621, 1996 8. Dalessio DJ: Cluster headache and cranial neuralgias. In Symposium of the 11th Annual Practicing Physician’s Approach to the Difficult Headache, February 20, 1998 9. Demaerel P, Boelaert I, Wilms G, et al: The role of cranial computed tomography in the diagnostic work-up of headache. Headache 36:347-348, 1996 10. Day J, Raskin N: Thunderclap headache: Symptom of unruptured cerebral aneurysm. Lancet 2:1247-1249, 1986 11. De Benedittis GD, Lorenzetti A, Sina C, et al: Magnetic resonance imaging in migraine and tension-type headache. Headache 35:264-268,1995 12. de Bruijn SFTM, Stam J, Kappelle LJ: Thunderclap headache as first symptom of cerebral venous sinus thrombosis. Lancet 348:1623-1625, 1996 13. Detailed Diagnoses and Procedures, National Hospital Discharge Survey, 1990. DHHS publication PHS 92-1774, Series 13, Hyattsville, MD, US Department of Health and Human Services, 1992 14. Diamond S Diagnosing and Managing Headaches. Professional Communications, 1998, pp 55-80 15. Duarte J, Sempere AP, Delgado JA, et al: Headache of recent onset in adults: A prospective population-based study. Acta Neurol Scand 94:67-70, 1996 16. Fontanarosa PB: Recognition of subarachnoid hemorrhage. Ann Emerg Med 18:11991205, 1989 17. Frishberg BM: The utility of neuroimaging in the evaluation of headache in patients with normal neurologic examinations. Neurology 44:1191-1197, 1994 18. Granella F, DAlessandro RD, Manzoni GC, et al: International Headache Society classification: Interobserver reliability in the diagnosis of primary headaches. Cephalalgia 14:16-20, 1994 19. Green C, Lumpkin JR, Baker FJ: Association between unsuspected carbon monoxide exposure and headache. 12:244-245,1983 20. Hale WE, May FE, Marks RG, et al: Headache in the elderly: An evaluation of risk factors. Headache 27272-276, 1987 21. Holloway RG, Kieburtz KD: Headache and the human immunodeficiency virus type 1 infection. Headache 35:245-255, 1995 22. Inagawa T, Hirano A: Autopsy study of unruptured incidental intracranial aneurysm. Surg Neurol34:361-365, 1990 23. Kahn CE, Sanders GD, Lyons EA, et al: Computed tomography for nontraumatic headache: Current utilization and cost-effectiveness. Can Assoc Radiol J 44:189-193, 1993 24. Levine HL: Diagnosis and treatment of sinonasal disease causing headache. In Symposium of The 11 Annual Practicing Physician’s Approach to the Difficult Headache, February 20, 1998 25. Liedo A, Calandre L, Martinez-Menendez B, et al: Acute headache of recent onset and subarachnoid hemorrhage: A prospective study. Headache March 1994;34:1724 26. Lipton RB, Feraru ER, Weiss G, et al: Headache in HIV-1 related disorders. Headache 311518-522, 1991 27. Lipton RB, Mazer C, Newman LC, et al: Sumatriptan relieves migrainelike headaches associated with carbon monoxide exposure. Headache 37392-395, 1997 28. Matsumura K, Matsuda M, Handa J, et a1 Magnetic resonance imaging with aneurysmal subarachnoid hemorrhage: Comparison with computed tomography scan. Surg Neurol3471-78, 1990 29. Merikangas KR, Frances A Development of diagnostic criteria for headache syndromes: Lessons from psychiatry. Cephalalgia 13(Suppl 12):34-38, 1993 30. Mitchell CS, Osborn RE, Grosskreutz SR Computed tomography in the headache patient: Is routine evaluation really necessary? Headache 33:82-86, 1993 31. Ogawa T, Inugami A, Fujita H, et al: MR diagnosis of subacute and chronic subarachnoid hemorrhage: Comparison with CT. Am J Radiol 165:1257-1262, 1995 32. Ogawa T, Inugami A, Shimosegawa E, et al: Subarachnoid hemorrhage: Evaluation with MR imaging. Radiology 186345-351, 1993 33. Olesen J: Headache Classification Committee of the International Headache Society,

EVALUATION OF THE PATIENT WITH NONTRAUMATIC HEADACHE

34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.

45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

151

Classification and Diagnostic Criteria for Headache Disorders, Cranial Neuralgia, and Facial Pain. Cephalalgia S(supp1 7):l-96, 1988 Ostergaard J R Headache as a warning symptom of impending aneurysmal subarachnoid hemorrhage. Cephalalgia 11:53-55, 1991 Osterhaus JT, Gutterman DL, Plachetka J R Healthcare resource and lost labor costs of migraine headache in the United States. Pharmacoeconomics 2:67-76,1992 Quality Standards Subcommittee of the American Academy of Neurology: Practice parameter: The utility of neuroimaging in the evaluation of headache in patients with normal neurologic examinations (summary statement). Neurology 44:1353-1354, 1994 Ramirez-Lassepas M, Espinosa CE, Cicero JJ, et al: Predictors of intracranial pathologic findings in patients who seek emergency care because of headache. Arch Neurol 54~1506-1509,1997 Raps EC, Rogers JD, Galetta SL, et al: The clinical spectrum of unruptured intracranial aneurysms. Arch Neurol50:265-268, 1993 Rasmussen BK, Jensen R, Schroll M, et al: Epidemiology of headache in a general population-a prevalence study. J Clin Epidemiol44:1147-1157, 1991 Ronkainen A, Vanninen R, Hernesniemi J: Familial aneurysms. Headache Quarterly 933638, 1998 Rorke LB, Schut L Introductory survey of pediatric brain tumors. In McLaurin RL (ed): Pediatric Neurosurgery, ed 2. Philadelphia, WB Saunders, 1989, pp 335-337 Rothner A D Headaches in childhood and adolescents. Semin Pediatr Neurol 2:lOl177, 1995 Sames TA, Storrow AB, Finkelstein JA: Sensitivity of new-generation computed tomography in subarachnoid hemorrhage. Acad Emerg Med 3:16-20, 1996 Schievink WI, Schaid DJ, Rogers HM, et al: Familial aneurysmal subarachnoid hemorrhage: A community-based study. J Neurosurg 83:426-429, 1995 Schwartz BS, et al: Epidemiology of tension-type headache. JAMA 279:381-383, 1998 Seymour JJ, Moscati RM, Jehle DV Response of headaches to nonnarcotic analgesics resulting in missed intracranial hemorrhage. Am J Emerg Med 13:43-45, 1995 Sidman R, Connolly E, Lemke T Subarachnoid hemorrhage diagnosis: Lumbar puncture is still needed when the computed tomography scan is normal. Acad Emerg Med 31827-831, 1996 Silberstein S D Evaluation and emergency treatment of headache. Headache 32396407, 1992 Silberstein SD, Marcelis J: Headache associated with changes in intracranial pressure. Headache 3284-94,1992 Silverberg E, Lubera J: Cancer statistics. Cancer 36:9-23, 1986 Singer EJ, Zorilla C, Fahy-Chandon B, et al: Painful symptoms reported by ambulatory HIV-infected men in a longitudinal study. Pain 5415-19, 1993 Smith R Diagnosing headaches. Hosp Med 33:2642, 1997 Snyder H, Robinson K, Shah D, et al: Signs and symptoms of patients with brain tumors presenting to the emergency department. J Emerg Med 11:253-258, 1993 Solomon S, Lipton R Criteria for the diagnosis of migraine in clinical practice. Headache 313386387, 1991 Stewart WF, Lipton RB, Celentano DD, et al: Prevalence of migraine headache in the United States: Relation to age, income, race, and other sociodemographic factors. JAMA 267:6449, 1992 Tomsak RL: Opthalmologic aspects of headache. Med Clin North Am 75:3, 693-705, 1991 van der Wee N, Rinkel GJE, Hasan D, et al: Detection of subarachnoid haemorrhage on early CT Is lumbar puncture still needed after a negative scan? J Neurol Neurosurg Psychiatry 58:357-359, 1995 van Gijn J, van Dongen KJ: The time course of aneurysmal haemorrhage on computed tomograms. Neuroradiology 23:153-156, 1982 Vazquez-Barquero A, Ibanez FJ, Herrera S, et al: Isolated headache as the presenting clinical manifestation of intracranial tumors: A prospective study. Cephalalgia 14:270272, 1994

152

FIELD & WANG

60. Vespa P: Acute presentation and early intensive care of acute aneurysmal subarachnoid hemorrhage. J Stroke Cerebrovasc Dis 6:4,230-234, 1997 61. Winner P, Wasiewski W, Gladstein J, et al: Multicenter prospective evaluation of proposed pediatric migraine: Revisions to the international headache society criteria. Headache 37545-548, 1997 62. Wijdicks E, Kerkoff H, Van Gijn J: Long-term followup of 71 Patients with thunderclap headache mimicking SAH. Lancet 2:68-70,1988

Address reprint requests to Adam G. Field, MD Division of Emergency Medicine Northwestern University School of Medicine Chicago, IL 60611